DEMOGRAPHIC RESEARCH
VOLUME 30, ARTICLE 55, PAGES 1527−1560
PUBLISHED 15 MAY 2014
http://www.demographic-research.org/Volumes/Vol30/55/
DOI: 10.4054/DemRes.2014.30.55
Research Article
Income pooling strategies among cohabiting and
married couples: A comparative perspective
Nicole Hiekel
Aart C. Liefbroer
Anne-Rigt Poortman
This publication is part of the Special Collection on “New Relationships from
a Comparative Perspective,” organized by Guest Editors Anne-Rigt Poortman
and Belinda Hewitt.
2014 Nicole Hiekel, Aart C. Liefbroer & Anne-Rigt Poortman.
This open-access work is published under the terms of the Creative Commons
Attribution NonCommercial License 2.0 Germany, which permits use,
reproduction & distribution in any medium for non-commercial purposes,
provided the original author(s) and source are given credit.
See http:// creativecommons.org/licenses/by-nc/2.0/de/
Table of Contents
1
Introduction
1528
2
2.1
2.2
1531
1531
1533
2.3
Theoretical background and hypotheses
Explanations focusing on selection processes
Explanations focusing on inherent differences of cohabitation and
marriage
The comparative setting
3
3.1
3.2
3.2
Data and methods
Data and sample
Measurements
Analytical approach
1538
1538
1539
1540
4
Results
1541
5
Conclusions
1548
6
Acknowledgements
1553
References
1554
1535
Demographic Research: Volume 30, Article 55
Research Article
Income pooling strategies among cohabiting and married couples:
A comparative perspective
Nicole Hiekel 1
Aart C. Liefbroer 2
Anne-Rigt Poortman 3
Abstract
BACKGROUND
Studies explaining why cohabiters are more likely to keep money separate than spouses
have mainly focused on selection processes, without taking into account the
heterogeneity within both union types in levels of commitment. Cross-national studies
are rare and have predominantly included Northern and Western European countries,
the United States, and Canada.
OBJECTIVE
This study explains the higher likelihood of cohabiters to keep income separate by
selection as well as commitment factors and explores country differences, including
countries from Central and Eastern Europe.
METHODS
Using data from the Generations and Gender Surveys of Bulgaria, France, Georgia,
Germany, Romania, and Russia, N=41,456 cohabiting and married individuals are
studied. Binary logistic regression models of the likelihood that respondents keep
money separate are calculated.
RESULTS
Across countries, higher education, female labor market participation, both partners
being employed, short union duration, absence of joint children, presence of separation
thoughts, and (for cohabiters) a lack of marital intentions are the most persistent
1
Netherlands Interdisciplinary Demographic Institute (NIDI), the Netherlands. University of Groningen, the
Netherlands. Correspondence to: Netherlands Interdisciplinary Demographic Institute (NIDI), the
Netherlands. P.O. Box 11650. 2502 AR, The Hague, the Netherlands. Telephone: +31(0)70 3565200.
Fax: +31 (0)70 3647187. E-Mail: [email protected].
2
Netherlands Interdisciplinary Demographic Institute (NIDI), the Netherlands. Faculty of Social Sciences,
VU University Amsterdam, the Netherlands. University of Groningen, the Netherlands.
3
Faculty of Social and Behavioral Sciences, Utrecht University, the Netherlands.
http://www.demographic-research.org
1527
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
correlates of keeping money separate. Differences between cohabiters and married
couples are reduced when selection and commitment are taken into account, but are still
significant. Cross-national variation in the effect of cohabitation on keeping separate
purses is persistent.
CONCLUSION
Different money management strategies of cohabiters and spouses can be explained to
some extent by selection processes and inherent differences in the level of commitment
within cohabitation and marriage. Countries also differ in the socio-economic context
and norms concerning the way intimate relationships are organized which might lead to
persistent differences in the way cohabiting and married couples organize their income.
1. Introduction
Income pooling strategies, namely, whether couples have a common purse or keep their
money apart, constitute an important aspect of a co-resident couple’s everyday life,
regardless of whether the partners are married or not. Income pooling strategies reflect
how individuals try to resolve the conflict between their commitment towards the
partner on the one hand and the maintenance of individual autonomy on the other hand.
Whether and how cohabiting and married couples differ with respect to income pooling
strategies is a research topic of growing relevance because cohabitation is becoming
more common, often preceding or replacing marriage (Kiernan 2001), and national
governments aim at regulating its practice. If cohabiters keep money separate more
frequently and if the number of cohabiters continues to increase, the popular
assumption of policy makers that households share all of their financial assets would be
challenged.
A consistent finding in existing research about income pooling strategies of
cohabiting and married couples has been that cohabiters are more likely to keep money
separate than their married counterparts (Ashby and Burgoyne 2008; Bradatan and
Kulcsar 2008; Elizabeth 2001; Lyngstad, Noack and Tufte 2011; Oropesa, Landale and
Kenkre 2003; Vogler, Brockmann and Wiggins 2006; Winkler 1997). The majority of
these studies focused on the role of selection. It has been suggested that cohabiters
differ from married couples before they enter cohabitation (Axinn and Thornton 1992;
van de Kaa 1993). Hence, individuals with certain characteristics are both effectively
selected into cohabitation as well as into individual money management. Our first
research question therefore is: To what extent do cohabiting and married couples differ
in the manner in which they manage money and how much of this variation is due to
selection into one of the two union types?
1528
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
Selection may, however, only be part of the story. Differences in the manner in
which money is managed could also result from inherent differences between
cohabitation and marriage; for instance, in the level of interpersonal commitment
(Rhoades, Stanley and Markman 2010; Stanley, Whitton and Markman 2004). It has
been argued that the most important difference between cohabitation and marriage is –
besides the legal differences – that cohabiters face a higher level of insecurity about the
relationship’s future, because cohabitation lacks strong institutional and normative rules
as well as the public affirmation of the marriage vow (Brines and Joyner 1999; Cherlin
2004; Winkler 1997). Cohabiting partners are thus expected to be less committed to
each other, resulting in a lower likelihood of income pooling. Commitment within
cohabitation and marriage, however, may vary. First, the economic lives of couples are
likely to get more intertwined with time; hence, long-term unions have a larger
tendency to pool economic resources than unions that have lasted for a shorter period of
time. Second, the presence of joint biological children in the household might indicate a
joint investment in the relationship that increases interdependence and solidarity
between partners and might reduce the difference between cohabitation and marriage
(Seltzer 2004). Third, doubts about the long-term stability of the union might
discourage individuals from striving for more financial interdependency. Our second
research question therefore is: To what extent are the differences between cohabiters’
and spouses’ income pooling strategies reduced when the level of interpersonal
commitment is taken into account?
A number of studies have shown that cohabiting relationships in which marriage
plans are present are not qualitatively different from marriages (Brown and Booth 1996;
Wiik, Bernhardt and Noack 2009). Marital intentions might be a sign of high
commitment, which reduces the perceived risk of income pooling. But the absence of
marital intentions does not necessarily imply low commitment to the union, as
cohabiters might view their union as an alternative for marriage. Also, the experience of
premarital cohabitation has been argued as signaling lower commitment to the
relationship, as partners have not been sure enough to marry directly (Forste and Tanfer
1996). But in many countries premarital cohabitation has become an increasingly
common way to start a co-resident union (Liefbroer and Billari 2010).
Barely any prior research has compared income pooling strategies of cohabiting
and married couples across different cultural and institutional settings. Most studies
have focused either on the differences between spouses and cohabiters in one country
(Ashby and Burgoyne 2008; Elizabeth 2001; Lyngstad, Noack and Tufte 2011;
Oropesa, Landale and Kenkre 2003; Vogler, Brockmann and Wiggins 2006; Winkler
1997) or compared married couples across countries (Lauer and Yodanis 2011; Yodanis
and Lauer 2007). The only two existing studies that have compared married and
cohabiting couples with regard to their income pooling strategies cross-nationally have
http://www.demographic-research.org
1529
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
found persistent differences between cohabiters and spouses in their income pooling
strategies after controlling for a limited set of selection and commitment factors, but do
not identify variation in the effect of cohabitation on money management strategies
across countries representing very different welfare state regimes (Hamplova and Le
Bourdais 2009; Heimdal and Houseknecht 2003).
Studying income pooling strategies of cohabiters and spouses in different social
and cultural contexts increases our understanding of the interplay between institutional
context, the selection into different union types, the meaning attached to cohabitation,
and the way intimate relationships are organized. First, when cohabitation is marginal,
the cohabiting population is more likely to consist of an overrepresentation of
individuals with specific characteristics that attach one predominant meaning to
cohabitation.
Second, countries might vary with regard to the commitment that is involved in
cohabitation. In Western European countries where most unions start by unmarried
cohabitation the level of commitment involved in these unions might vary largely
across individuals. Whereas some unions might dissolve relatively soon, for others the
difference between cohabitation and marriage might be blurry, as they view their union
as a long-term alternative to marriage and are very committed to their relationship. In
Eastern European countries where cohabitation is marginal and social approval of it is
low, the commitment involved in cohabiting unions might generally be higher, as
cohabitation tends to be short-lived and quickly converted into marriage.
Third, countries might not only vary with regard to selection into and commitment
within marriage, but also in cultural and social values associated with cohabitation and
marriage. Sociologists have posited an individualization of intimate relationships
occurring in contemporary Western societies, in which partners increasingly value
individual autonomy and self-realization. In order to maintain individual autonomy and
the ease of leaving a union that is no longer considered self-fulfilling, pooling income
might be avoided within individualized unions, leading to an increased likelihood of
individual money management (Lauer and Yodanis 2011).
Thus, our third research question is: How do countries differ in the association
between union type and income pooling strategies? More specifically, do we find
evidence that the country context translates differently in selection processes into
cohabitation and the level of commitment involved in cohabitation, and do we find
indications of different norms concerning money management across Europe that lead
to persistent cross-national differences in cohabiters’ and spouses’ income pooling
strategies, even when taking selection and commitment into account? Using data from
the Generations and Gender Surveys (GGS), we analyzed data from countries that differ
widely in the prevalence and institutional context of cohabitation, including not only
1530
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
two Western European countries, France and Germany, but also countries located in
Central and Eastern Europe: Bulgaria, Georgia, Romania, and Russia.
2. Theoretical background and hypotheses
The existing literature offers overwhelming evidence that cohabiters are more likely to
opt for independent money management than the married, although contrasting
explanations for these differences are proposed. Our first hypothesis concerns the
replication of the finding that cohabiters are more likely to keep money separate than
married individuals (Hypothesis 1). The two main explanations for this difference focus
on the role of selection processes and on inherent differences between marriage and
cohabitation.
2.1 Explanations focusing on selection processes
The selection argument implies that the same set of factors leads individuals to prefer
cohabitation and to opt for individual money management. First, age might be
associated with both union type choice and income pooling strategy. Cohabiters are on
average younger than married individuals. Young adults are more likely to be
economically dependent; for instance, on income provided by parents or study grants.
Thus they might not yet consider merging their income with their co-resident partner, or
at least might keep some income separate.
Second, selection may occur on the basis of cultural characteristics. Within the
theoretical framework of the Second Demographic Transition the increasing popularity
of cohabitation has been argued as resulting from a change in values and attitudes
concerning family life in a broader sense (Surkyn and Lesthaeghe 2004). The highly
educated have been considered as at the vanguard of this value change. Although at
later stages of the transition the educational gradient may have become smaller as larger
parts of the population enter into cohabitation, highly educated individuals have been
found to be more progressive in their value orientation, less in favor of marriage, and
more likely to cohabit (Kiernan 2000; Manting 1996). Individuals who value
individualism and personal autonomy are also more likely to have a preference for
separate purses in order to maintain financial independence (Elizabeth 2001) or to
simply facilitate decision-making on individual expenditure (Ashby and Burgoyne
2008). Several studies have reported that the higher educated are more likely to keep
their money separate (Hamplova and Le Bourdais 2009; Lyngstad, Noack and Tufte
http://www.demographic-research.org
1531
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
2011), but see Treas (1993) for contradictory evidence for African-American married
couples in the US.
It has also been argued that opting for unmarried cohabitation might be a sign of
social deprivation. In times of growing globalization and economic crisis the most
disadvantaged social strata of society might be more affected by the negative outcomes
of decreasing job stability and wage protection such as unemployment, job insecurity,
and economic uncertainty, and more likely to opt for cohabitation as a second best
option when marriage is not (yet) feasible (Blossfeld, Mills and Klijzing 2005;
McDonald 2006; Sobotka and Toulemon 2008).
Religiousness is another cultural resource on the basis of which this type of
selection might occur. Religious people hold more collectivistic and traditional values
that oppose unmarried cohabitation (Kiernan 2000; Manting 1996; Thornton, Axinn and
Xie 2007). Although no empirical evidence is available, it could well be that religious
people also prefer income pooling because this highlights their unique relationship.
Selection may also occur on the basis of the division of paid labor within the
couple. It has been found that couples in which the female partner is strongly attached
to the labor market are overrepresented among cohabiters (Kiernan 2000). Female labor
force participation and female earnings have been found to be positively associated with
independent money management as well (Elizabeth 2001; Hamplova and Le Bourdais
2009). Whereas the traditional male-breadwinner model requires that the employed
partner compensate the partner who specializes in home work by pooling economic
resources, the pre-eminence of specialization is reduced when both partners contribute
to the household income. Moreover, the bargaining power of the female partner in the
way the relationship is structured is enhanced when she contributes to the household
income.
Earlier life-course experiences could also influence union formation processes and
the manner in which relationships are organized (Guzzo 2006; Liefbroer, Gerritsen and
De Jong Gierveld 1994). Individuals who have experienced a divorce have been found
to be more likely to cohabit (Bumpass and Lu 2000). At the same time the experience
of a divorce might result in a reluctance to pool resources (Burgoyne and Morison
1997; Heimdal and Houseknecht 2003) as, for instance, previously married respondents
might have financial obligations towards their former partner, or children from a
previous relationship which might make them less willing to pool income with the
current partner (Burgoyne and Morison 1997). The presence of children from prior
unions in the household might increase the odds of independent money management, as
the step-parent might not want to pay for the child brought into the relationship or
might want to protect the biological parent’s alimony entitlement from his or her former
partner.
1532
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
If selection is at play, the observed differences between cohabiters and spouses
with regard to their income pooling strategies would be spurious, because they result
from differences in the background characteristics addressed above. In summary, our
second hypothesis reads: The association between union type and income pooling
strategy results from selection processes that shape both the preference for
cohabitation and for independent money management (Hypothesis 2).
2.2 Explanations focusing on inherent differences of cohabitation and marriage
In contrast to theoretical approaches that focus on selection, some authors have
emphasized that differences in the money management strategies of cohabiters and
spouses might also be explained by inherent differences between cohabitation and
marriage (Brines and Joyner 1999; Brown and Booth 1996; Nock 1995; Poortman and
Mills 2012). Often, differences in the commitment of married and cohabiting unions
have been highlighted. Marriage is a symbol of long-term commitment because it is
highly institutionalized (Nock 1995). Marriage is a public affirmation that implies
specific norms, obligations, and formal ties, and leads to “enforceable trust” (Cherlin
2004). Both financially and socially, marriage implies higher expectations and exit
costs than unmarried cohabitation. Married partners are therefore expected to be more
committed to each other than cohabiters. These differences in commitment are in turn
expected to translate into differences in the ways in which married and cohabiting
couples manage their money. Cohabiters face lower exit costs from their cohabiting
relationship, are more uncertain about the stability of the relationship due to the shorter
time horizon, and are confronted with more ambiguity regarding social expectations
about what it means to be a cohabiting partner. The uncertainty about the persistence
and seriousness of the relationship makes it risky for many cohabiters to pool their
income (Treas 1993), especially when joint property is not legally protected in the case
of separation. These differences have been argued as leading cohabiters to opt for
independent money management (Winkler 1997).
Although the level of commitment may generally be lower in cohabitations than in
marriages it is important to realize that the extent to which these union types differ in
their level of commitment might depend on specific characteristics of these unions.
First, union duration might matter. Cohabiting unions increasingly last longer (Sobotka
and Toulemon 2008). The economic lives of couples most likely get more intertwined
with time; hence, long-term unions have a stronger tendency to pool economic
resources than unions that have lasted a shorter period of time. Treas (1993) has
described this as a way to reduce transaction costs within households. Long-term
cohabiters might thus be as likely as married individuals to pool income with their
http://www.demographic-research.org
1533
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
partners. The difference between spouses’ and cohabiters’ money management
strategies might be thus largely due to the overrepresentation of short unions in
cohabiting relationships.
Second, across Europe, but also the United States and Canada, the increase in nonmarital childbearing is largely due to increasing births to cohabiting couples (Kiernan
2001; Raley 2001). A growing number of cohabiters have joint biological children. The
presence of joint children can represent increasing commitment to the union, as it
strengthens the bond between the parents (Seltzer 2000). In turn, children increase the
need of a couple to coordinate financial affairs. Moreover, a child increases the amount
of work within a household, which might require at least temporary specialization of
one partner in unpaid labor. This suggests that couples that share responsibilities for a
dependent household member are more likely to pool income in order to compensate
for specialization. The empirical evidence that cohabiters with joint biological children
differ less from married couples in their money management strategies (Hamplova and
Le Bourdais 2009; Lyngstad, Noack and Tufte 2011; Vogler, Brockmann and Wiggins
2006) suggests that the observed difference in the income pooling strategies in
cohabitation and marriage could be driven at least to some extent by the smaller
proportion of parents among cohabiting couples.
Third, not all married unions are per se more committed than cohabitations. The
level of relationship satisfaction has previously been found to be associated with
income pooling strategies (Hamplova and Le Bourdais 2009). Individuals who thought
about separation are less committed to their unions, which in turn might discourage
income pooling. Cohabiters have lower exit costs from their unions and might be
overrepresented among those considering a separation from their partner.
From these considerations on the role of commitment we derive a third hypothesis:
The differences in income pooling strategies of cohabiters and the married will be
reduced once the level of interpersonal commitment in both union types is taken into
account (Hypothesis 3).
A final aspect that may influence income pooling both within cohabitation and
marriage is whether cohabitation is linked to marriage or not. Some cohabiters plan to
marry while others do not. It is widely assumed that cohabiters with marriage plans are
more committed to their relationship than cohabiters who lack such intentions.
Cohabiters envisaging marriage have been found to be more likely to pool income than
cohabiters who do not have marriage plans (Brown and Booth 1996; Hamplova and Le
Bourdais 2009; Lyngstad, Noack and Tufte 2011). Some spouses cohabit before they
get married; others marry straight away. When separate purses are more likely during
cohabitation, married couples who cohabited before marriage might also be more likely
to continue the money management strategy employed during cohabitation.
1534
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
So far, no study on income pooling strategies of cohabiting and married couples
has taken the heterogeneity of both union types into account (but see Poortman and
Mills 2012 for other arrangements). We distinguish between those (a) cohabiting
without marital intentions, (b) cohabiting with marital intentions, (c) married with
premarital cohabitation, and (d) married without premarital cohabitation. We expect
these four union types to be differently associated with income pooling, suggesting a
hierarchical order. We expect income pooling to be least likely to occur among
cohabiters without plans to get married, followed by cohabiters who intend to marry.
Individuals who have cohabited with their partner before getting married are expected
to be more likely to pool income than both types of cohabiters, and, finally, income
pooling is expected to be most likely observed among married couples who married
directly (Hypothesis 4).
2.3 The comparative setting
Previous research has mainly focused on the income pooling strategies of couples in
Northern and Western Europe, the United States, and Canada. Much less attention has
been paid to countries in Central and Eastern Europe – countries that differ in many
ways and where significant societal, political, and demographic changes have taken
place during the last few decades. The present study examines income pooling
strategies of cohabiters and married individuals, comparing countries across Western
and Eastern Europe (Bulgaria, France, Georgia, Germany, Romania, and Russia), which
allows exploring general and context-specific differences between cohabitation and
marriage. Ultimately, this study helps us to understand the diversity of cohabitation
across Europe. In the following we briefly contextually situate the countries in our
study with regard to the prevalence and level of institutionalization of cohabitation. We
then discuss three reasons why we expect cross-national variation in the effect of union
type on money management: selection, commitment, and level of individualization in
cohabitation and marriage.
Northern European countries have been characterized as the forerunners in the
societal diffusion of cohabitation and have been referred to as “cohabitation land”
(Syltevik 2010: 448). Western European countries have been considered as following
suit. In Western Europe cohabitation has replaced direct marriage as the start of a union
for the vast majority of recent birth cohorts (own calculations based on GGS data, not
shown). In France cohabiters have been able to register their partnership since 1999,
when the “Loi sur la Concubinage et le Pacte Civil de Solidarité” (PACS) was
promulgated. In legal terms a registered partnership is largely similar to civil marriage.
Around half of all first births occur within cohabitation (Perelli-Harris et al. 2012).
http://www.demographic-research.org
1535
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
Cohabitation is part of the transition to adulthood in Germany as well. Several
institutional constraints and economic incentives encourage cohabiters to marry, and
marriage and parenthood are strongly linked. A registered partnership
(Lebensgemeinschaft) as an alternative to civil marriage is available to homosexual but
not heterosexual couples.
Central and Eastern European (CEE) countries have been classified at an earlier
stage of the diffusion process of cohabitation. Countries like Bulgaria, Georgia,
Romania, and Russia have been characterized as contexts where cohabitation is
marginal, practiced by a subpopulation with particular characteristics (Heuveline and
Timberlake 2004; Kiernan 2001). Nevertheless, substantial differences have been found
across countries in this region with regard to the prevalence of cohabitation and its
function in the childbearing process (Sobotka 2003). For instance, whereas 41% of
Georgian women born between 1971 and 1980 were cohabiting at the birth of their first
child, not more than 12% of their Romanian counterparts did so (own calculations
based on GGS data, not shown). But the predominant role of cohabitation in the union
career also differs across countries. For instance, in Russia divorced individuals as well
as urban residents are largely overrepresented among cohabiters (Zakharov 2008),
whereas in Bulgaria premarital cohabitation has a long tradition, where, particularly in
rural areas, cohabiters move in together at the moment of engagement, often into the
house of one set of parents (Hoem and Kostova 2008).
The extent to which cohabiting and married couples differ in their income pooling
strategies might vary between countries for at least three reasons. First, differences in
the prevalence of cohabitation across countries might lead to different selection
processes into cohabitation, as the above-mentioned examples illustrate. As a result of
the different composition of cohabiters across countries, differences in the income
pooling strategies of cohabiters and spouses might vary as well. In some countries only
a minority cohabits: in other countries virtually everyone experiences periods of
cohabitation. In countries where only a minority cohabits those who cohabit might
exhibit very different behavior from the married, and consequently likewise with regard
to income pooling strategies. These initially large differences between cohabiters and
spouses with regard to income pooling are expected to be strongly reduced after
selection factors are accounted for. In countries where cohabitation is widespread
selection into cohabitation is lower and differences in the income pooling strategies of
cohabiters and the married are expected to be smaller, and not substantially reduced by
taking selection factors into account.
Second, a contrasting hypothesis on cross-national differences in cohabiters’ and
spouses’ income pooling strategies can be derived when examining cross-national
differences in the commitment involved in cohabitation. When cohabitation is less
widespread, it constitutes a deviant behavior. Often, such a context is characterized by
1536
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
higher levels of religiousness and moral conservatism (Gault-Sherman and Draper
2012) and, consequently, strong societal pressure to get married (Lehrer 2004). This
may influence how cohabiters view their union, how stable the cohabitation is, and, if
cohabiters marry, how fast they do so. When cohabitation is marginal and stigmatized
cohabitation is likely to be short-lived and quickly transformed into marriage. In
countries where the decision to get married is interpreted as a consequence of high
interpersonal commitment we would expect commitment within cohabitation to be
higher. It has to be noted, however, that the aspiration of marriage can also indicate a
perceived absence of alternatives to marriage and thus an expression of conformism
(Coast 2009). If cohabiters in Central and Eastern Europe enter cohabitation
predominantly with the aspiration to marry soon they might adopt marriage-like
patterns of income organization right from the start of their union. At the country level
we would thus expect to find smaller differences between cohabiters’ and spouses’
money management. In contexts where cohabitation is more prevalent and accepted,
premarital cohabitation is viewed as a phase in the ‘normal’ life course and a significant
proportion of cohabiters might not marry at all, either because their union dissolves
rather than transforming into marriage— a process called “weeding” (Klijzing 1992:
53), or they consider their union as a permanent alternative to marriage. Thus there
might be great variation at the individual level in the commitment that is involved in
cohabitation. In addition, in these contexts the legal protection of non-marital unions is
often higher than in countries where cohabitation is rare (Perelli-Harris and Sánchez
Gassen 2012; Poortman and Mills 2012). If income pooling strategies reflect the
commitment of the union, differences in the money-management strategies of
cohabiters and spouses might be larger among Western Europeans. Accounting for the
variation in the level of commitment may, however, also reduce these differences most
strongly in these countries.
Finally, country differences might arise from differences in the socio-cultural
context, independently from differences in selection processes and levels of
commitment. It has been argued that individualization affects the nature of intimate
relationships and how couples manage their money. As the level of individualization
varies across countries it could be that in countries where the level of individualization
is high, intimate relationships are primarily entered into for the sake of satisfaction
derived from being with that partner rather than for the social recognition or economic
advantage gained by being in a partnership, and relationships that no longer provide
these benefits are easily dissolved (Giddens 1992). Building on new institutional
economics, which situates economic action into its institutional context, Lauer and
Yodanis (2011), in a cross-national comparative study on spousal income organization,
show that the spread of individualized marriage in a country influences spouses’ money
management strategies. In order to maintain individual autonomy and to ease leaving
http://www.demographic-research.org
1537
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
the union, married couples in individualized contexts are more likely to avoid income
pooling. Sociologists have argued that the individualization of intimate relationships is
increasingly a feature of Western societies (Beck and Beck-Gernsheim 1990). Keeping
income separate as an expression of individualistic values thus might be more likely in
Western Europe than in Eastern Europe. Cohabitation—even more than marriage—
might constitute an expression of the valorization of individual autonomy, and that the
relationship is entered into and maintained as long as it is considered self-fulfilling and
satisfying. Thus it could be that differences between cohabiters’ and spouses’ moneymanagement strategies are greater in Western Europe than in Eastern Europe.
Nevertheless, given that the majority of marriages in Western Europe nowadays are
preceded by a period of unmarried cohabitation, it could also be that spouses continue
the money management strategies employed during cohabitation, which would lead to
smaller differences between cohabiters and the married in Western Europe, compared to
Eastern Europe.
In sum, selection variables are expected to more strongly explain the effect of
union type on income pooling in Central and Eastern Europe, as cohabiters in these
countries are considered a more selective group than their Western European
counterparts. Commitment variables are expected to strongly reduce the differences
between cohabitation and marriage in Western Europe, as cohabiters in these countries
are expected to be more diverse with regard to the level of interpersonal commitment.
Persistent country differences might indicate that cross-national differences in the level
of individualization of intimate relationships exist.
3. Data and methods
3.1 Data and sample
We analyzed data from six countries that participated in the Generations and Gender
Surveys (GGS) that were collected in 2005 and 2006. The GGS is a panel survey of a
nationally representative sample of the 18–79 year-old resident population in each
participating country (Vikat et al. 2007). To date, Wave 1 data of 15 countries has been
collected. Six of them include a comparable measurement of the money management
strategy employed within the couple. These countries are Bulgaria, France, Georgia,
Germany, Romania, and Russia. The overall size of the main samples differs by country
but in most cases is about 10,000 respondents. Data were usually collected by a
computer-assisted personal interview (CAPI), and in some countries by a paper and
pencil interview (PAPI). The overall response rates vary between 49.7% in Russia and
78.2% in Bulgaria. Our analytical sample contains individuals who were either married
1538
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
or cohabited unmarried, with a heterosexual partner. Individuals who did not provide
information on their organization of household income (n=157) or who reported to have
an “other” manner in which they managed their money (n=474) were excluded from
further analysis. Moreover, we excluded respondents who did not report whether they
are married to their current partner (n=55), as well as cohabiters who did not report
whether they have intentions to marry within three years (n=114). The analyses are
based on a final sample of 36,407 (88%) married individuals and 5,049 (12%)
cohabiters.
3.2 Measurements
Income pooling. Our dependent variable – the extent to which income is pooled –
distinguished couples that kept their income at least partly separate from those that
pooled all their income. In the GGS respondents were asked:
“How do you and your partner/spouse organize your household income?”
1 – I manage all the money and give my partner/spouse his/ her share
2 – My partner/spouse manages all the money and gives me my share
3 – We pool all the money and each takes out what we need
4 – We pool some of the money and keep the rest separate
5 – We each keep our own money separate
6 – Other (Comment: only for coding, not presented in the card itself)
In line with previous studies (Heimdal and Houseknecht 2003; Vogler, Brockmann
and Wiggins 2006) we distinguished individuals who kept their income completely or
partly separate (responses 4 and 5) from those who pooled all their income (responses
1, 2, and 3). Pooling income is considered as the reference category.
Union type. Respondents who were married and live together with their partner are
distinguished from those who shared a household with a partner to whom they were not
married (coded as 1).
Selection factors. We included three age groups in the analysis: 18 to 35 years old, 36 to
55 years old, and 56 to 79 years old. The data provide an internationally comparable
measure of education attainment using the International Standard Classification of
Education ISCED (UNESCO 2006). We distinguished three levels: 1=primary and
lower secondary education, 2=upper secondary and post-secondary non-university
education, and 3=all levels of university education. The data include the self-reported
activity status of both partners. Four categories of the division of paid labor were
distinguished: 1=only the male partner is employed or self-employed, 2=only the
http://www.demographic-research.org
1539
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
female partner is employed or self-employed, 3=both are employed or self-employed,
4=none of them is employed or self-employed. Religiousness was measured as a
combination of religious denomination and the frequency of visiting religious services
(with the categories: less than once a year, once a year, less than once a month, once a
month, less than once a week, once a week, daily). We created a scale ranging from
1=no denomination or with denomination but visiting religious services less than once a
year, to 6=with denomination, going daily to church. A higher value indicated a higher
level of religiousness.
Life-course events. Dummy variables were created indicating whether the respondent
had ever been married with a former partner (coded as 1) and whether at least one
stepchild or biological child with a former partner and below age 18 was living in the
household.
Commitment factors. The union duration has been measured in years between the date
of interview and the date the couple started living in the same household. A dummy
variable was created that takes the value 1 when the respondent reported at least one
joint biological child younger than 18 years old with the current partner and living in
the household. To measure separation thoughts, respondents were asked “Over the past
12 months, have you thought about breaking up your relationship?” The answer
categories were 0=no and 1=yes.
Heterogeneity within cohabitation and marriage. Marital intentions of cohabiters were
ascertained by asking whether they intended to marry their partner within the next three
years. Respondents who answered 1=yes or 2=probably yes were considered as having
marriage plans. Those who responded 3=probably no or 4=no were treated as having no
marital intentions. Among spouses a distinction was made between those who
cohabited before marriage and those who married directly.
3.3 Analytical approach
First, descriptive statistics are discussed. These results provide insight into how
cohabiters and married individuals across countries differ in terms of selection and
commitment measures as well as in income pooling strategies. Within-country
differences between cohabiters and married have been tested for statistical significance.
We explicitly state whenever an observed difference does not reach statistical
significance. Next, our hypotheses are tested by estimating binomial logistic regression
1540
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
models for each country separately, because the small number of countries does not
allow the use of multi-level analysis (Hox 2010).
Item non-response is hardly problematic, because most of the variables had an
item-non-response rate of less than 1%. The question on separation thoughts has a high
item non-response of around 30% in both Georgia (for unknown reasons) and France
(as interviewers were instructed to skip the question on relationship assessment when
the respondent’s partner was present at the interview). In the case of continuous
variables with missing data, such as partnership duration and religiousness, we
performed value imputation using the Stata 10 “impute” command for each country
separately. In the case of categorical variables with missing data we created an
additional category, “no response”. For instance, respondents without separation
thoughts have been compared to respondents with separation thoughts as well as those
with a non-response.
4. Results
Table 1 shows for cohabiting and married individuals by country the percent
distribution of the variables included in our analysis. In CEE countries pooling all
income is by far the most common manner in which couples manage their money,
regardless of whether they are married or not. With the exception of Georgia and
Bulgaria where married and cohabiting couples do not differ significantly from each
other in the way they organize their income, independent money management is clearly
more frequently reported by cohabiting couples than by the married. Germany and
France differ from the CEE countries in two respects. First, the share of non-poolers
among married couples is higher (12% and 15%, respectively) than in Georgia,
Romania, and Russia. Second, the proportion of non-poolers among cohabiting couples
is much larger than among married couples. More than half of all cohabiting couples
keep at least some of their money separate (51% and 58%, respectively).
Countries vary in the prevalence of cohabitation. It is particularly uncommon in
Romania where approximately 5% of all co-resident unions are cohabitations. In
Georgia, Russia, Bulgaria, and Germany, the proportion of cohabiters varies between
10% and 15%. In France one fifth of all co-resident unions are cohabitations. More than
half of all cohabiting individuals in Bulgaria, Russia, Germany, and France intend to
marry their partner within three years. The vast majority of Georgian and Romanian
cohabiters reports marital intentions for the near future (78% and 67%, respectively).
Whereas around 50% of all currently married couples in each country experienced
premarital cohabitation, 17% of the Romanian and 32% of the Russian married couples
cohabited before they got married.
http://www.demographic-research.org
1541
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
Table 1:
Percent distribution of income management and characteristics of
the cohabiting and the married by country (N=41,456)
Bulgaria
coh
mar
Georgia
coh
Romania
mar
coh
Russia
mar
coh
mar
France
coh
Germany
mar
coh
mar
Income management
All income pooled
83.7
85.6
90.8
92.0
85.4
93.1
84.9
93.8
42.0
85.2
49.0
88.5
At least some income kept separate
16.3
14.4
9.2
8.0
14.6
6.9
15.1
6.2
58.0
14.8
51.0
11.5
Prevalence of cohabitation
Cohabiting among all coresident unions
10.2
13.6
5.3
14.7
20.4
12.3
Cohabiting without marital intentions
56.4
21.7
32.9
51.8
59.4
54.1
Cohabiting with marital intentions
43.6
78.3
67.1
48.2
40.6
Types of cohabitation/marriage
45.9
Married directly
45.6
52.9
83.0
67.5
51.1
49.4
Married after cohabitation
54.4
47.1
17.0
32.5
48.9
50.5
Selection variables
18–35 years old
68.8
23.3
49.7
17.6
37.4
16.6
43.3
24.8
50.5
13.4
50.8
12.8
36–55 years old
25.9
46.1
41.8
50.6
40.5
45.9
41.2
47.9
41.1
45.0
35.9
49.5
5.3
30.7
8.5
31.8
22.1
37.6
15.6
27.3
8.3
41.6
13.4
37.7
Primary education
50.2
26.4
11.4
12.3
53.4
37.3
14.5
12.3
19.8
35.0
17.1
11.3
Secondary education
38.5
51.2
60.8
59.5
39.9
52.7
56.6
47.7
47.5
39.9
58.6
60.0
Tertiary education
11.4
22.3
27.8
28.2
6.8
10.1
28.9
40.0
32.7
25.1
24.2
28.8
Only male partner employed
23.0
13.6
46.3
37.1
31.5
18.8
22.2
16.7
18.7
14.7
16.3
21.9
Only female partner employed
13.7
12.0
7.1
9.2
8.8
8.1
10.4
9.3
7.2
8.1
11.7
7.4
Both partners employed
32.4
44.5
21.9
23.6
31.5
37.3
48.3
46.4
64.7
45.1
53.2
43.1
No partner employed
31.0
29.9
24.6
30.0
28.2
25.9
19.2
27.6
9.4
32.0
18.9
27.6
2.4
2.6
2.9
2.7
3.3
3.8
1.9
1.9
1.3
1.8
1.6
2.5
21.9
2.8
6.7
1.3
37.6
4.5
48.7
12.2
19.6
7.3
21.4
6.0
10.2
1.2
2.2
0.2
13.1
1.4
24.3
3.6
11.9
2.0
10.7
5.9
7.8
24.5
10.7
25.0
10.7
26.3
8.1
22.4
8.1
26.7
6.7
25.1
56+ years old
Mean score religiousness scale (1–10)
Divorced
Child(ren) from prior unions below age 18
in household
Commitment variables
Mean union duration (in years)
Child(ren) from current union below age
18 in household
56.7
50.1
73.2
50.4
35.8
41.1
26.0
49.9
45.1
42.0
25.4
38.6
With separation thoughts
5.6
2.2
2.9
1.8
6.3
1.8
26.9
13.8
14.1
4.7
13.5
3.8
TOTAL NUMBER OF OBSERVATIONS
845
7,420 5,430
857
444 8,003
945 5,501 1,211 4,723
747 5,330
Note: Generations and Gender Surveys 2005/06.
Cohabiters in all countries are on average younger than married individuals;
cohabiters in CEE countries are in general older than cohabiters in Germany and
France. In all countries except Georgia the educational attainment of married and
cohabiting respondents differs significantly. In Romania and Bulgaria most cohabiters
1542
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
have only completed a low level of educational attainment (around 50%), whereas
married respondents are more likely to be highly educated. In Russia and Germany
these differences are smaller. French cohabiters, by contrast, are on average higher
educated than married individuals. Regarding the division of paid labor, we find that
more than half of French and German cohabiters live in unions in which both partners
are employed. The male breadwinner model is more prevalent in Central and Eastern
Europe, and in particular among cohabiters in Georgia and Romania. Both cohabiting
and married respondents in Romania on average score highest on the religiousness
scale, whereas Russian respondents, regardless of union status, and German cohabiters
are the least religious. Apart from Russia and Georgia where the differences do not
reach statistical significance, cohabiters are less religious than married individuals. In
all countries cohabiters report much more frequently that they have been previously
married. Although this proportion varies strongly across countries and is highest in
Russia and Romania (48% and 38%, respectively), the difference between married and
cohabiting individuals is remarkably large in all countries. Living together with children
below age 18 from prior unions is much more common among cohabiters and
particularly prevalent in Russia. In all countries the average partnership duration of
cohabiting unions is shorter than that of marriages. The difference between cohabiters
and spouses is largest in Germany, with 6 years for cohabiting versus approximately 25
years for married couples. Between a quarter (Germany, Russia) and 73% (Georgia) of
all cohabiting unions involve at least one joint biological child with the partner. Neither
married nor cohabiting individuals are likely to have thought about breaking up with
their partner during the last year. Among those who did cohabiters are largely
overrepresented, in particular in Germany and France, but also in Bulgaria, Georgia,
and Romania. Cohabiters are overrepresented among those with separation thoughts in
Russia as well. The peculiarity of Russia is that these kinds of thoughts are relatively
prevalent among married respondents as well.
In order to replicate previous findings and test Hypothesis 1, that cohabiters are
more likely to keep money separate, we conducted a logistic regression analysis with
independent money management as the dependent variable and union type as the only
independent variable, for each country separately (results not shown). With the
exception of Bulgaria and Georgia where the effect of union type on income pooling
does not reach statistical significance, cohabiters are more likely than married
respondents to opt for independent money management. Against our expectation that
differences between cohabiters and spouses are initially larger in CEE countries where
cohabitation is marginal, we find German and French cohabiters to have 8.0 times
higher odds of keeping money separate, whereas in Romania and Russia cohabiters’
odds of independent money management are increased by 2.7 and 2.3 times,
respectively.
http://www.demographic-research.org
1543
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
In Hypothesis 2 we assumed that the variation in income pooling strategies of
cohabiting and married individuals would be strongly reduced after controlling for
selection into cohabitation, as the differences between cohabiters’ and spouses’ income
organization would derive from individual characteristics associated with the entry into
cohabitation. Table 2 shows the results of a logistic regression analysis of the
association between union type and income pooling when selection processes into
cohabitation and independent money management are controlled for.
We do indeed find that keeping income separate is associated with variables that
are considered to select individuals into cohabitation, although not in all countries.
Individuals in the oldest age group are most likely to keep money separate in CEE
countries, but the effect reaches statistical significance only in Georgia and Russia. This
finding contradicts our assumptions and might be counter-intuitive. It is, however, in
line with previous research, and has been explained as a strategic response of older
individuals to country-specific inheritance tax regulations and medical care payments
(Heimdal and Houseknecht 2003; Lyngstad, Noack and Tufte 2011; Treas 1993). In
France and Germany the age affect is reversed, but as we will show below the effect
turns positive and significant when we include measurements of interpersonal
commitment. In line with prior studies we find that university-educated individuals are
more likely to keep money separate. In accordance with our theoretical expectation,
strong labor market participation of both partners and a female-breadwinner model
increases the likelihood of keeping separate purses. Surprisingly, we find religious
people to be slightly more likely to opt for separate purses in Georgia and Russia. In
line with our assumptions we also find previously married respondents to be more
likely to pursue individual money management. The presence of children from prior
unions in the household decreases the odds of separate purses only in Germany. We
thus find little empirical evidence that individuals avoid pooling resources when
children from prior unions are living in the household.
When selection variables are included in the model the effect of cohabitation on
keeping money separate becomes statistically significant in Bulgaria and Georgia. In
Romania and Russia the effect of union type remains virtually unchanged, whereas in
Germany and France the differences between cohabiters and the married are smaller
when selection is taken into account. Despite differences in the characteristics of
cohabiters in Western Europe compared to Central and Eastern Europe, our findings
suggest that selection processes into independent money management operate in similar
ways across countries. Against our expectation that selection factors would explain
more of the variation in CEE countries where cohabitation is less widespread, we find
that considering selection into cohabitation reveals differences in money management
strategies between cohabitation and marriage in Bulgaria and Georgia, but hardly
accounts for variation in two other Central and Eastern European countries. By contrast,
1544
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
taking selection into account explains some of the variation in income pooling strategies
by union type in Germany and France. Across Europe the effect of union type on
money management remains large. The odds ratio of separate purses among cohabiters
ranges between 1.3 in Georgia and 7.0 in Germany, after controlling for selection
processes.
Table 2:
Summary of logistic regression Analysis (eB) for variables predicting
independent money management by union type for cohabiters
(n=5,049) and the married (n=36,407), controlling for selection
processes
Bulgaria
Georgia
Romania
Russia
France
Germany
Union type (ref. married)
Cohabiting
1.60***
1.32**
2.45***
2.74***
5.92***
7.03***
36–55 years old
1.07
0.97
1.02
0.93
0.71***
0.90
56+ years old
1.24†
1.31**
1.21
1.13**
0.75**
0.95
Age (ref. 18–35 years)
Education (ref. primary education)
Secondary education
1.74***
1.01
1.23**
1.23
1.52***
1.17
Tertiary education
2.64***
1.40**
1.50**
1.69**
3.37***
1.83***
1.30**
2.64***
1.58**
1.41†
2.14***
1.34†
Both employed
1.49***
2.40***
2.02***
1.59**
2.00***
2.27***
None employed
0.98
2.57***
1.24
0.88
0.99
0.97
Religiousness (cont.)
1.00
1.07**
1.05
1.12**
0.97
0.98
0.88
1.40**
1.21
2.52***
2.04***
Division of labor (ref. male employed)
Female employed
Marriage history (ref. not divorced)
Divorced
0.87
Fertility history (ref. no child from prior union)
Child prior union
0.88
1.90
0.69
0.77
1.08
0.89
Constant
0.07***
0.03***
0.03***
0.03***
0.08***
0.07***
χ2
Pseudo R2
df
-3346.6
0.03
11
-1721.7
0.03
-2170.5
-1658.8
0.02
11
0.04
11
-2599.2
0.20
11
-2327.8
0.14
11
Note: Generations and Gender Surveys 2005/06. †p< .1 *p < .05, **p < .01, ***p < .001.
http://www.demographic-research.org
1545
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
In Hypothesis 3 we assumed that differences in the income organization of
cohabiters and the married would derive from variation in the level of interpersonal
commitment, both within cohabitation and marriage. In order to test that hypothesis we
added to the model union duration, presence of joint biological children in the
household, and separation thoughts. The results are presented in Table 3.
Table 3:
Summary of logistic regression Analysis (eB) for variables predicting
independent money management by union type for cohabiters
(n=5,049) and the married (n=36,407), controlling for selection
processes (omitted from table) and level of commitment
Bulgaria
Georgia
Romania
Russia
France
Germany
Union type (ref. married)
Cohabiting
1.60***
1.29†
2.32***
2.11***
3.93***
4.62***
Union duration
0.99**
0.99
1.00
0.98***
0.95***
0.97***
0.73**
0.89
0.64***
0.62***
Fertility history current union (ref. no joint child)
Joint child
0.65***
0.88
Evaluation relationship (ref: no separation thoughts)
Sep. thoughts
2.83***
1.74***
4.55***
2.44***
2.22***
1.52**
Constant
0.11***
0.04***
0.03***
0.03***
0.17***
0.13***
χ
2
Pseudo R2
df
-3314.2
0.04
14
-1716.2
0.04
14
-2139.3
0.03
14
-1624.7
0.06
14
-2515.7
0.22
14
-2295.6
0.15
14
Note: Generations and Gender Surveys 2005/06. †p< .1 *p < .05, **p < .01, ***p < .001.
Including the commitment measures decreases the effect of union type on money
management strategies in all countries. In Georgia differences between cohabiting and
married individuals remain only marginally statistically significant. The effects of the
selection variables remain virtually unchanged, with the exception of the effects of two
variables (results not shown). First, older age turns from a negative to a positive effect
in Germany and France, and thus becomes in line with findings from all other countries.
Second, the negative association between the presence of children from prior unions in
the household and separate purses reaches marginal statistical significance in Romania
and statistical significance at the .05 level in Germany. Likelihood ratio tests reveal that
controlling for interpersonal commitment significantly improved the model fit for each
of these countries.
1546
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
The longer the union duration, the lower the likelihood that a couple keeps income
separate. The effect of union duration is particularly strong in Western countries where
the effect of union type decreases strongly after partnership duration is controlled for
(step-wise models not reported but available upon request). As expected, joint
biological children significantly decrease the odds of separate purses in four out of six
countries, suggesting that a couple’s financial lives get more intertwined when partners
share the responsibility for a joint child. Separation thoughts turn out to be very
strongly linked to income pooling patterns in all countries. Respondents who
considered breaking up with their partner during the last year have a strongly increased
likelihood of keeping money separate.
Although including commitment factors reduces the cross-national variation in the
effect of cohabitation on independent money management it remains significant and
varies between 1.3 in Georgia and 4.6 in Germany.
Finally, in Hypothesis 4 we expected that taking into account cohabiters’ marriage
plans and married individuals’ cohabitation experience might explain differences in the
income organization of both union types. Table 4 presents the results of a regression
model that replaces the dichotomized union type variable by a fourfold typology in
order to test within-group heterogeneity.
In all countries both types of married individuals—those who married directly and
those who did so after cohabitation with their spouse—are less likely to keep money
separate than cohabiters. Only in Romania and France do we observe statistically
significant differences in the income pooling strategies of both groups of married
respondents. Whereas Romanian spouses who cohabited before they got married are
less likely to keep money separate than those who married directly, French excohabiters are—as hypothesized—more likely to keep money separate than those who
married directly. For the other countries these results indicate that income pooling
strategies within a marriage are not associated with premarital cohabitation. We thus
found little support for the assumption that premarital cohabitation is relevant for
income pooling strategies within marriage.
Apart from Bulgaria and Georgia where no or only marginal differences were
observed between any of the cohabitation and marriage types, we find that cohabiters
with marriage plans differ from the married couples less than cohabiters without such
plans. Cohabiters who do not have short-term marriage intentions have higher odds of
keeping separate at least some of their income than the ‘marriage-minded’ cohabiters.
Differences between both types of cohabiters are only statistically significant in Russia,
Germany, and France. We thus find some empirical support for within-group
heterogeneity among cohabiters with regard to income organization.
http://www.demographic-research.org
1547
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
Summary of logistic regression Analysis (eB) for variables predicting
independent money management by two types of cohabiters
(n=5,049) and two types of married (n=36,407), controlling for
selection processes and level of commitment (omitted from table)
Table 4:
Bulgaria
Union type (ref. married directly)
Married after
cohabitation
1.06
Cohabit with
marriage plans
1.38†
Cohabit without
marriage plans
1.35†
Constant
0.11***
χ
2
-3313.8
2
Pseudo R
df
0.04
16
Georgia
Romania
Russia
France
Germany
0.98
0.67**
1.13
1.47***
1.14
1.22
2.00**
1.74**
4.09***
4.16***
1.43
2.19**
2.72***
6.42***
5.89***
0.04***
0.04***
0.03***
0.13***
0.12***
-1715.9
0.04
16
-2134.5
-1621.6
0.04
16
0.06
16
-2502.9
0.13
16
-2292.2
0.15
16
Note: Generations and Gender Surveys 2005/06. †p< .1 *p < .05, **p < .01, ***p < .001.
5. Discussion
In this paper cohabiting and married couples were compared with regard to their
income pooling strategies across different countries. The aim was to examine the extent
to which cohabiters and the married differ in their income pooling strategies, and to
analyze whether the factors that account for these differences vary across countries. In
accordance with prior research (Elizabeth 2001; Hamplova and Le Bourdais 2009;
Heimdal and Houseknecht 2003; Lyngstad, Noack and Tufte 2011; Vogler, Brockmann
and Wiggins 2008) support was found for Hypothesis 1 that cohabiters are more
inclined towards independent money management in four of the six countries. We did
not find such differences in Georgia and Bulgaria. We can imagine two explanations.
First, cohabiting unions in these countries tend to be particularly short-lived and the
predominant exit from cohabitation is marriage rather than separation (own cohort
analysis of the cumulative incidence of marriage and separation as two competing
events based on GGS data, not shown). Thus it could be that cohabiters adapt income
pooling as the predominant way in which money is organized within marriage right
from the start of their union. Second, as we have discussed earlier, income pooling
might be an expression of economic constraints rather than an individual preference and
1548
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
couples might need to pool all their resources to make ends meet. Within our sample of
countries Georgia and Bulgaria have the lowest Gross Domestic Product (GDP) and are
particularly poor European countries (The World Bank 2012). It could thus be that
independent money management is not feasible for couples in these countries,
regardless of their marital status.
We examined the relevance of selection in explaining the differences between
cohabiters’ and spouses’ money management. Although several potential selection
factors were related to income pooling, these factors were hardly capable of explaining
the relationship between union type and income pooling strategy. We find that older
individuals, those with a university degree, unions in which the female or both partners
are highly attached to the labor market, and the previously divorced (in Western
Europe) to be more inclined to keep money separate. These results replicate findings for
other countries (Burgoyne and Morison 1997; Hamplova and Le Bourdais 2009;
Heimdal and Houseknecht 2003; Lyngstad, Noack and Tufte 2011; Treas 1993; Vogler,
Brockmann and Wiggins 2008). In all countries the effect of cohabitation on
independent money management remained considerable, implying that the selection
factors considered here were not explaining much of the differences in the income
pooling strategies of cohabiting and married couples, and if they did, they did so more
in contexts where cohabitation is more prevalent. Given the popularity of the selection
hypothesis, our findings suggest that selection cannot be the whole story and that it is
likely that there must be inherent differences between cohabitation and marriage that
influence how couples manage their money. Selection, however, might still play a role.
We could only include a limited number of selection factors that have been discussed in
the literature and our choice might not be exhaustive. Particularly for the Eastern
European countries very little is known about the selection processes into cohabitation,
and we might have missed factors attracting individuals to cohabitation and individual
money management in these societies.
Next, we examined the assumption that inherent differences between cohabitation
and marriage concerning the relationship itself explain this variation. We do indeed find
that differences in the interpersonal commitment of cohabiting and married couples are
associated with their money management strategies. Differences between cohabiting
and married couples were strongly reduced (particularly in Germany and France) after
controlling for union duration in four of the six countries. The longer a relationship
lasts, the more entangled the partners’ lives become and the more likely it becomes that
they opt for pooled economic resources. First, our results imply that the different
average union duration of cohabiting and married relationships accounts for a lot of the
variation in the money management within both union types. This is in line with the
findings of Lyngstad et al. (2011) for Norway. Most of the previous studies on
cohabiters’ money management, however, did not include union duration in their
http://www.demographic-research.org
1549
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
analysis (Hamplova and Le Bourdais 2009; Vogler, Brockmann and Wiggins 2008).
Second, these findings suggest that long-term cohabiters are likely to behave more
similarly to married couples in the economic organization of their household income.
This is interesting because it is widely assumed that long-term cohabiting unions are the
most ideologically inspired ones. If so, it suggests that there are institutional pressures
on these couples that make them pool their incomes (e.g., buying a house together).
Joint biological children below age 18 are associated with lower odds of keeping
money separate. This finding suggests that a couple’s financial lives get more
intertwined when they have joint responsibility for a biological child. This finding has
important implications, as an increasing number of children are born to cohabiting
parents. It suggests that the presence of children might become an increasingly crucial
determinant in the money management strategy of couples, regardless of the union type
they are in. Despite on-going changes in individuals’ living arrangements and family
context, parenthood encourages financial solidarity within a couple, whether the parents
are married or not.
In line with our expectations, separation thoughts were associated with a higher
likelihood of keeping income separate. We argued that thinking about breaking up
signifies low commitment, which in turn discourages individuals to pool their income.
However, some sort of reversed causality could be operative as well. The absence of a
joint pool could cause or signify a lower commitment to the partner and in turn lead to
thoughts about separation. Longitudinal data are needed to shed light on this issue.
In three of the six countries (Germany, France, Russia) cohabiters with marriage
plans are less likely to keep money separate than cohabiters without marital intentions,
a finding also reported by Lyngstad and colleagues (2011) for Norway. Cohabiters
anticipating marriage might already have a preference for joint finances or might start
adopting marriage-like behaviors after the decision to get married has been made. This
difference within the group of cohabiters stresses the importance of taking the
heterogeneity among cohabiters into account. Cohabiters anticipating marriage behave
more similarly to married couples than cohabiters not thinking about getting married.
We find little support for the hypothesis that married couples who cohabited before
their marriage are more likely to keep money separate during marriage than those who
married straight away. The only exception is France, where virtually all co-resident
unions start by unmarried cohabitation. Those who marry directly thus constitute a
selective group, and their high likelihood to have joint money management might be
related more to religiousness or conservatism than to different levels of interpersonal
commitment.
Although we were able to illustrate the existence of substantial heterogeneity in
income pooling strategies among both cohabiting and married couples the effect of
cohabitation on keeping money separate remained strong and significant. This finding
1550
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
could indicate that the institution of marriage induces a couple to attach a special
significance to their relationship that differs substantively from cohabitation. In contrast
to cohabitation, marriage is a social institution, legally regulated and surrounded by
clear social expectations about how a married couple should behave. The norms
concerning marriage also concern financial arrangements such as material support and
inheritance, and thus offer an institutional framework for mutual solidarity between
spouses. Hence, marriage seems to be strongly associated with joint finances in all
countries included in our study. This finding implies that marriage is such a strong
institution that it mainstreams prior cohabiters. Since the decision as to whether to pool
income has to be made by every couple at some point in the course of their relationship,
the availability of longitudinal data for future research would allow examining at what
point in the course of an intimate relationship the decision to pool income is being
made, as well as which couples change their money management over time.
Another limitation of this study is that we do not differentiate between partial
pooling and keeping all income separate. Burgoyne and colleagues (2007) identified
partial pooling as a distinct money management strategy, as it shares similarities with
both independent money management and complete pooling. Couples employing partial
pooling manage a significant proportion of the income independently, but both partners
have access to joint money for collective expenses. We acknowledge this important
distinction, which is particularly relevant in thinking about its consequences for access
to and control of financial resources within couples. Our data did not allow us to further
distinguish among those not pooling all of their resources as the number of observations
is too small.
Finally, we explored whether country differences in the effect of union type on
income pooling strategies were related to different selection processes into money
management strategies across countries or cross-national variation in the level of
commitment in cohabitation and marriage. We find that selection processes into
independent money operate in similar ways across countries, despite the cohabiting
populations differing in some characteristics. Therefore our results suggest that
selection explains little of the differences between cohabiters’ and spouses’ different
ways of managing money. Variation in commitment factors, however, cannot
exhaustively explain the variation in cohabiters’ and spouses’ income pooling strategies
in different contexts.
It might be that the money management strategy that a couple pursues is not only
the outcome of an individual preference, but also mirrors constraints. The socioeconomic context of cohabiters might thus play a role in how feasible it is to keep
money separate. Cohabiters in CEE countries are characterized by a lower level of
educational attainment and participation in the labor market, both compared to married
individuals in their country and compared to cohabiters in Western Europe. For them
http://www.demographic-research.org
1551
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
the question of whether money should be kept separate might simply not come up,
because these couples need all their money to keep things running, even if the couple
has ideological qualms about it. This explanation is supported by our finding that
differences in the money management of cohabiters and spouses in Central and Eastern
Europe are small to begin with, and hardly influenced by selection or commitment
variables. In prosperous countries where people earn more than they need for their basic
needs the idea of keeping at least some of the money separate in order to increase
individual autonomy might make sense.
Nevertheless, the level of commitment accounts better for the variation in Western
European countries than in Central and Eastern Europe. Another explanation of the
persistent country differences could thus be that the more normative cohabitation
becomes, the more diverse the cohabiting population in terms of commitment involved.
The persistence of country differences in the extent to which cohabiters are more likely
to keep money separate gives some indication of the notion that Western European
relationships are indeed more individualized, but also that cohabitation is a particularly
individualized union type. In Western Europe individualistic and postmodern values are
highly accepted. People in these countries might therefore abstain from income pooling
for ideological reasons such as maintaining their individual autonomy, even though
they are highly committed to their partner. Cohabitation, even more than marriage,
might be an expression of holding such individualized values. As a consequence levels
of commitment cannot exhaustively explain why cohabiters are more likely to keep
their money separate than married couples. Alternatively, it could be that even in
Western Europe, where cohabitation is common and accepted, keeping money separate
during cohabitation might be a norm itself. When common sense does not expect
cohabiters to pool economic resources, cohabiters might be less likely to do so. In
Central and Eastern Europe, by contrast, norms to pool income within a couple might
be particularly strong, regardless of whether the partners are married or not. When
keeping income separate is considered a particular deviant behavior it is not very likely
to occur.
We explored interesting country differences in the six countries we examined,
hereby opening promising avenues for future research. Analyzing countries separately
does not provide the empirically appropriate test of the effect of the institutional
context. When data on more countries is available it will be possible to directly test for
their impact on money management strategies multi-level models that include both
individual and country-level indicators.
1552
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
6. Acknowledgements
This research results from collaboration between members of the Relations-CrossNations (RCN)-network. RCN meetings have been supported by: (a) a grant from
Utrecht University within the University of California-Utrecht University Collaborative
Grant Program (2009), (b) an RFP-grant from the Population Association of America
(2010/11), (c) a grant from the Spanish Ministry of Science (2010), and (d) grants from
the Swiss National Science Foundation and FORS (2011).
http://www.demographic-research.org
1553
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
References
Ashby, K.J. and Burgoyne, C.B. (2008). Separate financial entities? Beyond categories
of money management. The Journal of Socio-Economics 37(2): 458–480.
doi:10.1016/j.socec.2006.12.035.
Axinn, W.G. and Thornton, A. (1992). The Relationship between cohabitation and
divorce – selectivity or causal influence. Demography 29(3): 357–374. doi:10.
2307/2061823.
Beck, U. and Beck-Gernsheim, E. (1990). Das ganz normale Chaos der Liebe.
Frankfurt am Main: Suhrkamp.
Blossfeld, H.-P., Mills, M., and Klijzing, E. (2005). Becoming an adult in uncertain
times: a 14-country comparison of the losers of globalization. In H.-P. Blossfeld,
M. Mills, and K. Kurz (eds.), Globalization, uncertainty and youth in society.
London: Routledge: 423–441.
Bradatan, C. and Kulcsar, L. (2008). Choosing between marriage and cohabitation:
Women's first union patterns in Hungary. Journal of Comparative Family
Studies 39(4): 491–507.
Brines, J. and Joyner, K. (1999). The ties that bind: Principles of cohesion in
cohabitation and marriage. American Sociological Review 64(3): 333–355.
doi:10.2307/2657490.
Brown, S.L. and Booth, A. (1996). Cohabitation versus marriage: A comparison of
relationship quality. Journal of Marriage and the Family 58(3): 668–678.
doi:10.2307/353727.
Bumpass, L. and Lu, H.-H. (2000). Trends in cohabitation and implications for
children's family contexts in the United States. Population Studies 54(1): 29–41.
doi:10.1080/713779060.
Burgoyne, C.B. and Morison, V. (1997). Money in remarriage: Keeping things simple –
and separate. The Sociological Review 45(3): 363–394. doi:10.1111/1467954X.00069.
Burgoyne, C.B., Reibstein, J., Edmunds, A., and Dolman, V. (2007). Money
management systems in early marriage: Factors influencing change and stability.
Journal of Economic Psychology 28(2): 214–228. doi:10.1016/j.joep.2006.
02.003.
1554
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
Cherlin, A.J. (2004). The deinstitutionalization of American marriage. Journal of
Marriage and Family 66(4): 848–861. doi:10.1111/j.0022-2445.2004.000
58.x.
Coast, E. (2009). Currently cohabiting: Relationship attitudes, expectations and
outcomes. In J. Stillwell, E. Coast, and D. Kneale (eds.), Fertility, living
arrangements, care and mobility: understanding population trends and
processes. Dordrecht: Springer: 105–125. doi:10.1007/978-1-4020-9682-2_6.
Elizabeth, V. (2001). Managing money, managing coupledom: a critical examination of
cohabitant's money management practices. The Sociological Review 49(3):
389–411. doi:10.1111/1467-954X.00338.
Forste, R. and Tanfer, K. (1996). Sexual exclusivity among dating, cohabiting, and
married women. Journal of Marriage and the Family 58(1): 33–47. doi:10.2307/
353375.
Gault-Sherman, M. and Draper, S. (2012). What will the neighbors think? The effect of
moral communities on cohabitation. Review of Religious Research 54(1): 45–67.
doi:10.1007/s13644-011-0039-9.
Giddens, A. (1992). The transformation of intimacy. Cambridge: Polity Press.
Guzzo, K.B. (2006). The relationship between life course events and union formation.
Social Science Research 35(2): 384–408. doi:10.1016/j.ssresearch.2005.06.002.
Hamplova, D. and Le Bourdais, C. (2009). One pot or two pot strategies? Income
pooling in married and unmarried households in comparative perspective.
Journal of Comparative Family Studies 40(3): 355–385.
Heimdal, K.R. and Houseknecht, S.K. (2003). Cohabiting and married couples' income
organization: Approaches in Sweden and the United States. Journal of Marriage
and Family 65(3): 525–538. doi:10.1111/j.1741-3737.2003.00525.x.
Heuveline, P. and Timberlake, J.M. (2004). The role of cohabitation in family
formation: The United States in comparative perspective. Journal of Marriage
and Family 66(5): 1214–1230. doi:10.1111/j.0022-2445.2004.00088.x.
http://www.demographic-research.org
1555
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
Hoem, J.M. and Kostova, D. (2008). Early traces of the Second Demographic
Transition in Bulgaria: A joint analysis of marital and non-marital union
formation, 1960–2004. Population Studies 62(3): 259–271. doi:10.1080/003247
20802313256.
Hox, J. (2010). Multilevel Analysis. Techniques and Applications. London: Routledge.
Kiernan, K. (2000). European perspectives on union formation. In: Waite, L., Bachrach,
C., Hindin, M., Thomson, E., and Thornton, A. (eds.). The ties that bind:
Perspectives on marriage and cohabitation. New York: Aldine Gruyter: 40–58.
Kiernan, K. (2001). The rise of cohabitation and childbearing outside marriage in
western Europe International Journal of Law, Policy and the Family 15(1):
1–21. doi:10.1093/lawfam/15.1.1.
Klijzing, E. (1992). 'Weeding' in the Netherlands: First union disruption among men
and women born between 1928 and 1965. European Sociological Review 8(1):
53–70.
Lauer, S. and Yodanis, C. (2011). Individualized marriage and the integration of
resources. Journal of Marriage and Family 73(3): 669–683. doi:10.1111/j.17413737.2011.00836.x.
Lehrer, E.L. (2004). Religion as a determinant of economic and demographic behavior
in the United States. Population and Development Review 30(4): 707–726.
doi:10.1111/j.1728-4457.2004.00038.x.
Liefbroer, A.C. and Billari, F.C. (2010). Bringing norms back in: A theoretical and
empirical discussion of their importance for understanding demographic
behaviour. Population, Space and Place 16(4): 287–305. doi:10.1002/psp.552.
Liefbroer, A.C., Gerritsen, L., and De Jong Gierveld, J. (1994). The influence of
intentions and life course factors on union formation behavior of young adults.
Journal of Marriage and the Family 56(1): 193–203. doi:10.2307/352713.
Lyngstad, T.H., Noack, T., and Tufte, P.A. (2011). Pooling of economic resources: A
comparison of Norwegian married and cohabiting couples. European
Sociological Review 27(5): 624–635. doi:10.1093/esr/jcq028.
Manting, D. (1996). The changing meaning of cohabitation and marriage. European
Sociological Review 12(1): 53–65. doi:10.1093/oxfordjournals.esr.a018177.
McDonald, P. (2006). Low fertility and the State: The efficacy of policy. Population
and Development Review 32(3): 485–510. doi:10.1111/j.1728-4457.2006.001
34.x.
1556
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
Nock, S.L. (1995). Commitment and dependency in marriage. Journal of Marriage and
the Family 57(2): 503–514. doi:10.2307/353702.
Oropesa, R.S., Landale, N., and Kenkre, T.S. (2003). Income allocation in marital and
cohabiting unions. Journal of Marriage and Family 65(4): 910–926.
doi:10.1111/j.1741-3737.2003.00910.x.
Perelli-Harris, B., Kreyenfeld, M., Sigle-Rushton, W., Keizer, R., Lappegård, T.,
Jasilioniene, A., Berghammer, C., and Di Guilio, P. (2012). Changes in union
status during the transition to parenthood in eleven European countries, 1970s to
early 2000s. Population Studies 66(2): 167–182. doi:10.1080/00324728.2012.
673004.
Perelli-Harris, B. and Sánchez Gassen, N. (2012). How similar are cohabitation and
marriage? Legal approaches to cohabitation across Western Europe. Population
and Development Review 38(3): 435–467. doi:10.1111/j.1728-4457.2012.005
11.x.
Poortman, A.-R. and Mills, M. (2012). Investments in marriage and cohabitation: The
role of legal and symbolic commitment. Journal of Marriage and Family 74(2):
357–376. doi:10.1111/j.1741-3737.2011.00954.x.
Raley, K.R. (2001). Increasing fertility in cohabiting unions: Evidence for the second
demographic transition in the United States? Demography 38(1): 59–66. doi:10.
1353/dem.2001.0008.
Rhoades, G.K., Stanley, S.M., and Markman, H.J. (2010). Should I stay or should I go?
Predicting dating relationship stability from four aspects of commitment.
Journal of Family Psychology 24(5): 543–550. doi:10.1037/a0021008.
Seltzer, J.A. (2000). Families formed outside of marriage. Journal of Marriage and the
Family 62(4): 1247–1268. doi:10.1111/j.1741-3737.2000.01247.x.
Seltzer, J.A. (2004). Cohabitation in the United States and Britain: Demography,
kinship, and the future. Journal of Marriage and Family 66(4): 921–928.
doi:10.1111/j.0022-2445.2004.00062.x.
Sobotka, T. (2003). Re-emerging diversity: Rapid fertility changes in Central and
Eastern Europe after the collapse of the communist regimes. Population-E
58(4–5): 451–486.
Sobotka, T. and Toulemon, L. (2008). Changing family and partnership behaviour:
Common trends and persistent diversity across Europe. Demographic Research
19(6): 85–138. doi:10.4054/DemRes.2008.19.6.
http://www.demographic-research.org
1557
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
Stanley, S.M., Whitton, S.W., and Markman, H.J. (2004). Maybe I do: Interpersonal
commitment and premarital or nonmarital cohabitation. Journal of Family Issues
25(4): 496–519. doi:10.1177/0192513X03257797.
Surkyn, J. and Lesthaeghe, R. (2004). Value orientation and the second demographic
transition (SDT) in Northern, Western and Southern Europe: An update.
Demographic Research S3(3): 45–86. doi:10.4054/DemRes.2004.S3.3.
Syltevik, L.J. (2010). Sense and sensibility: cohabitation in 'cohabitation land'.
Sociological Review 58(3): 444–462. doi:10.1111/j.1467-954X.2010.01932.x.
The World Bank (2012). World Development Report 2013: Jobs. Washington D.C.:
The World Bank.
Thornton, A., Axinn, W.G., and Xie, Y. (2007). Marriage and cohabitation. Chicago:
Chicago University Press. doi:10.7208/chicago/9780226798684.001.0001.
Treas, J. (1993). Money in the bank: Transaction costs and the economic organization
of marriage. American Sociological Review 58(5): 723–734. doi:10.2307/209
6283.
UNESCO (2006). International standard classification of education ISCED 1997.
van de Kaa, D.J. (1993). Second demographic transition revisited: Theories and
expectations. In G.C.N. Beets, H. van den Brekel, R. Cliquet, G. Dooghe, and J.
de Jong Gierveld (ed.), Population and family in the Low Countries 1993. Lisse:
Swets and Zeitungen: 81–126.
Vikat, A., Spéder, Z., Beets, G., Billari, F.C., Bühler, C., Désesquelles, A., Fokkema,
T., Hoem, J.M., MacDonald, A., Neyer, G., Pailhé, A., Pinnelli, A., and Solaz,
A. (2007). Generation and Gender Survey: Towards a better understanding of
relationships and processes in the life course. Demographic Research 17(14):
389–440. doi:10.4054/DemRes.2007.17.14.
Vogler, C., Brockmann, M., and Wiggins, R.D. (2006). Intimate relationships and
changing patterns of money management at the beginning of the twenty-first
century. British Journal of Sociology 57(3): 455–482. doi:10.1111/j.1468-4446.
2006.00120.x.
Vogler, C., Brockmann, M., and Wiggins, R.D. (2008). Managing money in new
heterosexual forms of intimate relationships. Journal of Socio-Economics 37(2):
552–576. doi:10.1016/j.socec.2006.12.039.
1558
http://www.demographic-research.org
Demographic Research: Volume 30, Article 55
Wiik, K.A., Bernhardt, E., and Noack, T. (2009). A study of commitment and
relationship quality in Sweden and Norway. Journal of Marriage and Family
71(3): 465–477. doi:10.1111/j.1741-3737.2009.00613.x.
Winkler, A.E. (1997). Economic decision-making by cohabitors: findings regarding
income pooling. Applied Economics 29(8): 1079–1090. doi:10.1080/000368497
326471.
Yodanis, C. and Lauer, S. (2007). Managing money in marriage: Multilevel and crossnational effects of the breadwinner role. Journal of Marriage and Family 69(5):
1307–1325. doi:10.1111/j.1741-3737.2007.00449.x.
Zakharov, S. (2008). Russian Federation: From the first to second demographic
transition. Demographic Research 19(24): 907–972. doi:10.4054/DemRes.2008.
19.24.
http://www.demographic-research.org
1559
Hiekel, Liefbroer & Poortman: Income pooling strategies among cohabiting and married couples
1560
http://www.demographic-research.org